The Study Of Re-Combinable Multi-Channel Patient Monitor Based On Associative Analysis Of Physiological Signal

A patient monitor is an instrument or a system designed to perform associated monitoring of physiological parameters and signal the alarms for the abnormal conditions. Traditionally, the functions and the physiological parameters monitored of a particular patient monitor are fixed, and each physiological parameter is analyzed individually. However, some diseases are related with more than one parameter simultaneously and hence the precise diagnosis of these diseases requires the associative analysis of multiple physiological parameters. According to such situation, a re-combinable patient monitor is introduced in this topic, which is based on the modular design and supports the hot-swap technology and can diagnose the diseases automatically. The implementation of the plug-in units and hot-swap-supported interface facilitates the medical staff to recombine the functions of the patient monitor, which expands the functions and improves the versatility and stability. Moreover, based on the associative analysis of multiple physiological parameters, this system has the capability of automatic diagnosis.The cerebral infarction is a disease which endangers the health of human beings, especially the middle-aged and older people. Since the mortality and disability rates of the cerebral infarction are high, this disease brings about serious harm to the patients and their families. In clinical practice, the cerebral infarction is diagnosed by computed tomography (CT) and magnetic resonance imaging (MRI) at present, which are expensive and poor real time. In this topic, an automatic diagnosis system of the cerebral infarction is designed on the basis of the feature detection using wavelet transformation and the associative analysis of electrocardiograph (ECG) signal and blood pressure signal. Moreover, this system can diagnose other diseases by installing the particular analysis programs.The major research works of this paper include:1. The modular design of the system. This patient monitor is based on the modular design. To diagnose the cerebral infarction, the electrocardiograph module, non-invasive blood pressure measurement module (NBP) and invasive blood pressure (IBP) measurement module are designed in this project. According to the actual requirement, other function modules can be installed flexibly.2. The implementation of hot-swap interface. The monitoring of the critical patient and surgical patients should not be interrupted. However, at times, the change of the function modules is necessary due to the requirement of monitoring other parameters or the fault in the existing modules. To satisfy this demand, a universal interface with hot-swap support is designed in this project, which makes possible changing the function modules without interrupting the work of the overall system and improves the stability of the system.3. The construction of the model of cerebral infarction in rats and the extraction of signals. Because of the similarity of the physiological characteristics between the rats and humans, a model of cerebral infarction in rats is constructed in this research. By analyzing the aberrances of physiological parameters of rats, the physiological parameters of humans are predicted and the analogical research of the diagnosis of human diseases is conducted. Therefore, the rats are used to substitute the human as the subjects in the experiments. In this research, a photochemical method is introduced to build the model of cerebral infarction, and the electrocardiograph signal and blood pressure signal are extracted.4. The de-noising and feature detection based on wavelet transformation. Most of the physiological signals are weak and contain considerable noises. By using the wavelet transformation, the noises and useful signals are disintegrated into different scales. The useful signals are distinguished from the noises by changing the wavelet coefficient. Moreover, the characteristic values of the electrocardiograph signal are detected by calculating the maximum value of the wavelet transformation.5. Establish the correlation model of the features of the electrocardiograph signal and blood pressure signal using the data mining technology. The clinical doctors’ observations indicate that the electrocardiograph signal and blood pressure signal of the patient would show abnormity simultaneously during the acute cerebral infarction period. The abnormity is reflected as the raised diastolic blood pressure and systolic blood pressure, the flat or inverted T-wave in cardiogram, the abnormity in S-T segment, prolonged Q-T interval and so forth. Based on a large number of experimental, the association rules model of the features of cardiogram and blood pressure of cerebral infarction is established using the associative analysis technology, which provides evidence for the diagnosis of diseases.